Sketching Dictionary Based Robust PCA in Large Matrices

Xingguo Li; Jarvis Haupt

doi:10.1109/IEEECONF44664.2019.9048695

Sketching Dictionary Based Robust PCA in Large Matrices

Xingguo Li, Jarvis Haupt

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper, our goal is to rapidly locating a few outliers and recover a low dimensional space spanned by inliers, with the particular interest when the outliers have known basis. A simple two-step approach is proposed based on a sketching procedure with theoretical guarantees for the performance. We show that exact identification of the outliers and recovery of the subspace can be achieved using sketched samples as few as the rank plus the number of outliers, ignoring logarithmic factors. This results in significant improvement on both sampling and computational efficiency. Comprehensive numerical experiments are provided to show the efficiency of our proposed method.

Original language	English (US)
Title of host publication	Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019
Editors	Michael B. Matthews
Publisher	IEEE Computer Society
Pages	702-706
Number of pages	5
ISBN (Electronic)	9781728143002
DOIs	https://doi.org/10.1109/IEEECONF44664.2019.9048695
State	Published - Nov 2019
Event	53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019 - Pacific Grove, United States Duration: Nov 3 2019 → Nov 6 2019

Publication series

Name	Conference Record - Asilomar Conference on Signals, Systems and Computers
Volume	2019-November
ISSN (Print)	1058-6393

Conference

Conference	53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019
Country/Territory	United States
City	Pacific Grove
Period	11/3/19 → 11/6/19

Bibliographical note

Funding Information:
The authors would like to graciously acknowledge support from DARPA Young Faculty Award, Grant No. N66001-14-1-4047

Publisher Copyright:
© 2019 IEEE.

Keywords

data sketching
dictionary
outlier identification
robust PCA

Access

10.1109/IEEECONF44664.2019.9048695

OpenUrl availability

Full text

Cite this

Li, X., & Haupt, J. (2019). Sketching Dictionary Based Robust PCA in Large Matrices. In M. B. Matthews (Ed.), Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019 (pp. 702-706). Article 9048695 (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2019-November). IEEE Computer Society. https://doi.org/10.1109/IEEECONF44664.2019.9048695

Sketching Dictionary Based Robust PCA in Large Matrices. / Li, Xingguo; Haupt, Jarvis.
Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019. ed. / Michael B. Matthews. IEEE Computer Society, 2019. p. 702-706 9048695 (Conference Record - Asilomar Conference on Signals, Systems and Computers; Vol. 2019-November).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Li, X & Haupt, J 2019, Sketching Dictionary Based Robust PCA in Large Matrices. in MB Matthews (ed.), Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019., 9048695, Conference Record - Asilomar Conference on Signals, Systems and Computers, vol. 2019-November, IEEE Computer Society, pp. 702-706, 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019, Pacific Grove, United States, 11/3/19. https://doi.org/10.1109/IEEECONF44664.2019.9048695

@inproceedings{3106b732f4784c14865e9927d7b8226b,

title = "Sketching Dictionary Based Robust PCA in Large Matrices",

abstract = "In this paper, our goal is to rapidly locating a few outliers and recover a low dimensional space spanned by inliers, with the particular interest when the outliers have known basis. A simple two-step approach is proposed based on a sketching procedure with theoretical guarantees for the performance. We show that exact identification of the outliers and recovery of the subspace can be achieved using sketched samples as few as the rank plus the number of outliers, ignoring logarithmic factors. This results in significant improvement on both sampling and computational efficiency. Comprehensive numerical experiments are provided to show the efficiency of our proposed method.",

keywords = "data sketching, dictionary, outlier identification, robust PCA",

author = "Xingguo Li and Jarvis Haupt",

note = "Funding Information: The authors would like to graciously acknowledge support from DARPA Young Faculty Award, Grant No. N66001-14-1-4047 Publisher Copyright: {\textcopyright} 2019 IEEE.; 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019 ; Conference date: 03-11-2019 Through 06-11-2019",

year = "2019",

month = nov,

doi = "10.1109/IEEECONF44664.2019.9048695",

language = "English (US)",

series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",

publisher = "IEEE Computer Society",

pages = "702--706",

editor = "Matthews, {Michael B.}",

booktitle = "Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019",

}

TY - GEN

T1 - Sketching Dictionary Based Robust PCA in Large Matrices

AU - Li, Xingguo

AU - Haupt, Jarvis

PY - 2019/11

Y1 - 2019/11

N2 - In this paper, our goal is to rapidly locating a few outliers and recover a low dimensional space spanned by inliers, with the particular interest when the outliers have known basis. A simple two-step approach is proposed based on a sketching procedure with theoretical guarantees for the performance. We show that exact identification of the outliers and recovery of the subspace can be achieved using sketched samples as few as the rank plus the number of outliers, ignoring logarithmic factors. This results in significant improvement on both sampling and computational efficiency. Comprehensive numerical experiments are provided to show the efficiency of our proposed method.

AB - In this paper, our goal is to rapidly locating a few outliers and recover a low dimensional space spanned by inliers, with the particular interest when the outliers have known basis. A simple two-step approach is proposed based on a sketching procedure with theoretical guarantees for the performance. We show that exact identification of the outliers and recovery of the subspace can be achieved using sketched samples as few as the rank plus the number of outliers, ignoring logarithmic factors. This results in significant improvement on both sampling and computational efficiency. Comprehensive numerical experiments are provided to show the efficiency of our proposed method.

KW - data sketching

KW - dictionary

KW - outlier identification

KW - robust PCA

UR - http://www.scopus.com/inward/record.url?scp=85083322117&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85083322117&partnerID=8YFLogxK

U2 - 10.1109/IEEECONF44664.2019.9048695

DO - 10.1109/IEEECONF44664.2019.9048695

M3 - Conference contribution

AN - SCOPUS:85083322117

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 702

EP - 706

BT - Conference Record - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019

A2 - Matthews, Michael B.

PB - IEEE Computer Society

T2 - 53rd Asilomar Conference on Circuits, Systems and Computers, ACSSC 2019

Y2 - 3 November 2019 through 6 November 2019

ER -

Sketching Dictionary Based Robust PCA in Large Matrices

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this